Re: [PATCH 06/18] chainlint.pl: validate test scripts in parallel

From: Eric Sunshine <sunshine@sunshineco.com>
Date: Nov 21, 2022, 14:07 UTC
Message-ID: <CAPig+cS3Ui=SFuRLPKKugT9RFvtUV3FmO23Wse_Rhih5hgbPmg@mail.gmail.com>
In-Reply-To: <221121.86tu2sbfh8.gmgdl@evledraar.gmail.com>

On Mon, Nov 21, 2022 at 8:32 AM Ævar Arnfjörð Bjarmason <avarab@gmail.com> wrote:

Show 12 quoted lines

> On Sun, Nov 20 2022, Eric Sunshine wrote:
> > Somehow Windows manages to be unbelievably slow no matter what. I
> > mentioned elsewhere (after you sent this) that I tested on a five or
> > six year old 8-core dual-boot machine. Booted to Linux, running a
> > single chainlint.pl invocation using all 8 cores to check all scripts
> > in the project took under 1 second walltime. The same machine booted
> > to Windows using all 8 cores took just under two minutes(!) walltime
> > for the single Perl invocation to check all scripts in the project.
>
> I'd be really interested in seeing e.g. the NYTProf output for that run,
> compared with that on *nix (if you could upload the HTML versions of
> both somewhere, even better).

Unfortunately, I no longer have access to that machine, or usable Windows in general. Of course, someone else with access to a dual-boot machine could generate such a report, but whether anyone will offer to do so is a different matter.

Show 16 quoted lines

> Maybe "chainlint.pl" is doing something odd, but this goes against the
> usual wisdom about what is and isn't slow in Perl on windows, as I
> understand it.
>
> I.e. process star-up etc. is slow there, and I/O's a bit slower, but
> once you're started up and e.g. slurping up all of those files & parsing
> them you're just running "perl-native" code.
>
> Which shouldn't be much slower at all. A perl compiled with ithreads is
> (last I checked) around 10-20% slower, and the Windows version is always
> compiled with that (it's needed for "fork" emulation).
>
> But most *nix versions are compiled with that too, and certainly the one
> you're using with "threads", so that's not the difference.
>
> So I suspect something odd's going on...

This is all my understanding, as well, which is why I was so surprised by the difference in speed. Aside from suspecting Windows I/O as the culprit, another obvious possible culprit would be whatever mechanism/primitives "ithreads" is using on Windows for locking/synchronizing and passing messages between threads. I wouldn't be surprised to learn that those mechanisms/primitives have very high overhead on that platform.

Show 6 quoted lines

> > Overall, I think Ævar's plan to parallelize linting via "make" is
> > probably the way to go.
>
> Yeah, but that seems to me to be orthagonal to why it's this slow on
> Windows, and if it is that wouldn't help much, except for incremental
> re-runs.

Oh, I didn't at all mean that `make` parallelism would be helpful on Windows; I can't imagine that it ever would be (though I could once again be wrong). What I meant was that `make` parallelism would be a nice improvement and simplification (of sorts), in general, considering that I've given up hope of ever seeing linting be speedy on Windows.